Strategies for Co-processing in

Refineries

Techno-economic & Refinery Impact

Analysis

NRELMike TalmadgeAvantika Singh

PNNLYuan JiangJalal Askander

February 12, 2020SCR Project Meeting, Richland, WA

2

Outline

• Introduction

• Refinery LP Model for Refinery Impact Analysis

• Co-Processing at Mild Hydrocracking Unit

• Co-Processing at Fluidized Catalytic Cracking Unit

• Discussions and Future Plan

3

Co-Processing Value to Bio-Refiner

Raw Insertion Co-Processing Product Blending

Conversion Hydroprocessing Separation

Crude Distillation Unit Upgrading Blending

Crude oil Straight run cuts Petroleum Intermediates On-spec Fuels

Biomass Bio-oilBiocrude

Hydrotreated Biofuel Fuel Blends

Cases (2016$) Catalytic Fast Pyrolysis (2000 dry tonne/day)

Waste Hydrothermal Liquefaction (110 dry tonne/day)

Reduction in MFSP ($/gge) 0.83 0.65MFSP w/o SCR ($/gge) 4.28 3.11

Potential Cost Saving at Bio-Refinery (upgrading at a standalone biorefinery vs an existing petroleum refinery)

Refinery Integration – Co-Processing

Co-processing intermediate bio-oil or biocrude at an existing petroleum refinery has the potential to significantly reduce CAPEX and therefore minimum fuel selling price (MFSP) of a bio-refinery.

least cost-effective

4

LP Model to Assess Refinery Impacts

The economic value of co-processing biomass intermediates can be evaluated by comparing the gross margins of a petroleum refiner before and after adding bio-oil/biocrude comparing the break-even value of bio-oil/biocrude to petroleum refiner and its minimum

selling price at bio-refinery.

A Bird’s Eye View of Full Refinery Linear Programming Optimization in Aspen PIMS

? Yields & Utilities ? Operating Severity ? Feed & Product Quality

Biogenic Carbon Diesel Production Less Sulfur Less Metals? Stability & Miscibility ? Acidity & Corrosion ? Oxygenates ? CO2 & H2O generation ? H2 Consumption

? Up- & Down-streams Operations

5

LP Model to Assess Refinery Impacts

• Mass balance and key properties of co-processing from experimental data

Model Inputs

• Process model and assay data of petroleum refinery from Aspen PIMS Gulf Coast Example

• Process constraints• Fuel specification• Feed and Product

Slates• Unit capacity

Model OutputsGross marginBreak-even valueBlending constraintsMass/Energy BalanceBasis for LCA

? Yields & Utilities ? Operating Severity ? Feed & Product Quality

Biogenic Carbon Diesel Production Less Sulfur Less Metals? Stability & Miscibility ? Acidity & Corrosion ? Oxygenates ? CO2 & H2O generation ? H2 Consumption

? Up- & Down-streams Operations

6

Refinery Analysis Inputs for SCR Project

+ +

Crude and product pricing data from OPIS by IHS Market.

ASTM finished fuel specifications are consistent with industry.

Bio-intermediate costs and refinery yields from SCR project experiments.

Fuel market projections from EIA, OPIS and ADOPT.

Refinery configuration and unit capacity basis from EIA.

7

Process Model for Mild Hydrocracker Co-Processing

• Preliminary Aspen Plus & TEA models have been developed for single-stage fixed-bed mild hydrocracker

• Detailed process model will be updated based on the coming co-processing experimental data

• Results from Aspen Plus model will be leveraged in the full refinery LP model

Assumptions in preliminary model• One-stage HDC for bio-oil• Similar yield as VGO HDC• O removed via H2O and CO2

• No combination effects (either deleterious or synergistic)

To be updated in new model• Yield• Gas phase • Combination effects• Two-stage HDC for biocrude

8

• Base-Delta Model in Refinery LP model

𝒚𝒚𝒊𝒊 = 𝒚𝒚𝒊𝒊𝒃𝒃𝒃𝒃𝒃𝒃𝒃𝒃 + �𝒃𝒃𝒌𝒌 𝒒𝒒𝒇𝒇,𝒌𝒌 − 𝒒𝒒𝒇𝒇,𝒌𝒌𝒃𝒃𝒃𝒃𝒃𝒃𝒃𝒃 + �𝒃𝒃𝒋𝒋 𝒛𝒛𝒋𝒋 − 𝒛𝒛𝒋𝒋𝒃𝒃𝒃𝒃𝒃𝒃𝒃𝒃

𝑦𝑦 = yield, product quality; 𝑞𝑞𝑓𝑓 = feedstock quality, z = operating condition, a = parameter ( ⁄∆𝑦𝑦 ∆𝑞𝑞𝑓𝑓, ⁄∆𝑦𝑦 ∆𝑧𝑧), 𝑓𝑓= capacity

• Base-Delta Model for Mild Hydrocracker𝒚𝒚 𝒒𝒒𝒇𝒇 𝒛𝒛

Hydrogen consumptionProduct Yield

H2S, CO2, H2O, Fuel Gas, C3, C4, Light Naphtha, Heavy Naphtha, Kerosene, Diesel, Unconverted Oil

Product QualitiesRequired by downstream upgrading, or fuel blending (i.e. Cetane Index)

Utility Consumption

Specific GravitySulfur, wt%50% ASTM Distillation Temp, oFBasic Nitrogen, ppmvOxygen, wt%

Feed RateConversion (%)Recycle (% of fresh feed)Catalyst Age (% used)

Subject to 𝑞𝑞𝑓𝑓,𝑘𝑘𝐿𝐿 ≤ 𝑞𝑞𝑓𝑓,𝑘𝑘 ≤ 𝑞𝑞𝑓𝑓,𝑘𝑘

𝑈𝑈 , 𝑧𝑧𝑗𝑗𝐿𝐿 ≤ 𝑧𝑧𝑗𝑗 ≤ 𝑧𝑧𝑗𝑗𝑈𝑈, 𝑓𝑓𝐿𝐿 ≤ 𝑓𝑓 ≤ 𝑓𝑓𝑈𝑈

LP structure has been modified in Aspen PIMS to adopt bio-oil/biocrude; Value of parameter a will be updated based on the coming experimental data and Aspen Plus model in Q3 &Q4

LP Model for Mild Hydrocracker Co-Processing

9

Pyrolysis Oil Co-Processing in FCC

• NREL is focused on developing full refinery optimization models with integrated co-processing of pyrolysis oil intermediates in the Fluid Catalytic Cracking unit.

• Initially building models based on Petrobras-NREL CRADA project (Pinho et al, 2017) and prior Strategic Analysis / TC Platform Analysis work.

• Model design methodology to enable quick incorporation of SCR project results.

10

FCC Co-Processing Analysis in FY2020

• Applied initial yield basis for FCC co-processing of pyrolysis oil from Petrobras-NREL CRADA.

• Improved yield model to allow user to assess pyrolysis oils of varying quality based on prior co-processing work in Strategic Analysis and TC Platform Analysis tasks.FCC Product X Yield = f (oxygen content, % FP oil in FCC feed, FCC operating conditions)

• Leveraging refinery impact and co-processing work from (1) Co-Optima ASSERT and (2) TC Platform Analysis.o Refinery configuration and unit capacity basis per region.o Variable crude and product pricing structure from OPIS. o Finished product specifications from ASTM international. o Fuel market projections from EIA, OPIS and ADOPT model.

Example of Gasoline Pricing Model

Example of Refinery Product Demand Projections

11

Preliminary FCC Co-Processing Analysis

• Develop understanding of the optimal co-processing strategy:

o Reduce bio-intermediate MSP by avoiding Biorefinery upgrading capital and operating costs.

o Determine bio-intermediate values to refineries through Aspen PIMS models.

o Valuation of high-value co-products from catalytic pyrolysis processes.

Preliminary example. Please do not cite or distribute.

12

• Optimized scenarios have maximum delta between “Total Value” and “Total Cost”.

• Refinery analysis can help direct R&D.

• Preliminary insights:o Maximize yields.

o Maximize co-product value.

o Analysis can help optimize. Preliminary example. Please do not cite or distribute.

Preliminary FCC Co-Processing Optimization

13

Next Steps for SCR Co-Processing Analysis

Oxygen Species Refinery Fuels LPG Products Distillate ProductsCO CO2 Water Dry Gas (C2-) Propane Propylene Cracked Naphtha

Coke i-Butane n-Butane Butenes Light Cycle Oil Resid Fuel Oil

• Experimental yield data Develop statistical yield models Incorporate yields into refinery LP models

Develop basis for process models Derive basis for environmental analysis

Graphs from TCS 2016, Chapel Hill, NC, November 1, 2016

14

• Feed and product properties Empirical physical property models Product allocation analysis Blending and fuel quality analysis

ASTM certification and spec development LCA / GHG analysis Inconsistencies in valuation / allocation

Example of insights developed from bio-oil physical property data.

Graphs from NREL TC Platform Analysis Q4 FY16 milestone

Next Steps for SCR Co-Processing Analysis

15

Downstream Refinery Impact Assessments

Assist in assessing co-processing topics outside of modeling scope:

• Blendstock quality impacts resulting from co-processing.

• Bio-intermediate logistics costs, challenges and constraints.

• Operational reliability impacts to downstream and ancillary unit operations:Managing low bio-oil thermal stability, managing immiscibility of bio-oils with refinery streams, acidity and corrosion potential, alkali and alkaline earth metals, unconverted oxygenated species, H2O, CO and CO2 in refinery light end / fuel gas streams, analysis methods for renewable allocations.

• Potential benefits to refiners from co-processing: Biofuels tax incentives, crude distillation capacity, FCC regenerator air blower capacity, FCC wet gas compressor, main fractionator flooding in FCC or Hydrocracker, reduced vanadium and nickel, low sulfur to sulfur plant and low nitrogen to sour water stripper.

Thank you!

Questions?NRELMike TalmadgeAvantika Singh

PNNLYuan JiangJalal Askander

This work was authored in part by the National Renewable Energy Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. Funding provided by the U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Bioenergy Technologies Office. The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government. The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes.

NREL/PR-5100-76131